[PATCH] x86/alternative: PAUSE is not a NOP

[Peter Zijlstra]

Subject: x86/alternative: PAUSE is not a NOP
From: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Date: Wed Jun 14 16:31:35 CEST 2023

While chasing ghosts, I did notice that optimize_nops() was replacing
'REP NOP' aka 'PAUSE' with NOP2. This is clearly not right.

Fixes: 6c480f222128 ("x86/alternative: Rewrite optimize_nops() some")
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Link: https://lore.kernel.org/linux-next/20230524130104.GR83892@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/
---
 arch/x86/kernel/alternative.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -169,9 +169,12 @@ void text_poke_early(void *addr, const v
  */
 static bool insn_is_nop(struct insn *insn)
 {
-	if (insn->opcode.bytes[0] == 0x90)
+	/* Anything NOP, but no REP NOP */
+	if (insn->opcode.bytes[0] == 0x90 &&
+	    (!insn->prefixes.nbytes || insn->prefixes.bytes[0] != 0xF3))
 		return true;
 
+	/* NOPL */
 	if (insn->opcode.bytes[0] == 0x0F && insn->opcode.bytes[1] == 0x1F)
 		return true;